Early Immunocastration of Male Pigs Effects on Physiology, Performance and Behaviour
Carl Brunius
Faculty of Natural Resources Department of Food Science Uppsala
Doctoral Thesis Swedish University of Agricultural Sciences Uppsala 2011
Acta Universitatis Agriculturae Sueciae 2011:84
Cover: Effects of early immunocastration on physiology, performance and behaviour. (Illustration: C. Brunius)
ISSN 1652-6880 ISBN 978-91-576-7628-3 © 2011 Carl Brunius, Uppsala Print: SLU Service/Repro, Uppsala 2011
Early immunocastration of male pigs –
Effects on physiology, performance and behaviour
Abstract The aim of this study was to investigate the effects of prepubertal or early pubertal vaccination against GnRH using Improvac® (Pfizer Ltd.) on boar taint, reproductive organs, anabolic hormones, cytochrome P450 enzymes, performance and behaviour. Crossbred male pigs (n=192) were randomly allocated to four groups: one group surgically castrated without anaesthesia, a second group receiving early vaccination (at ages 10 and 14 weeks), a third standard vaccinated group (at ages 16 and 20 weeks), and a fourth group of entire male pigs. After the second injection, antibody titres increased rapidly and testicular steroids decreased to the low levels of castrates. Reproductive organs were small in vaccinated pigs and smaller after early vaccination. Spermatogenesis and steroidogenesis were disrupted with a more severe, possibly irreversible effect after early vaccination. Oestradiol was suppressed for castrated and vaccinated pigs. IGF-1 was lowest for castrates and highest for entire male pigs, with vaccinated pigs at an intermediate level. Hepatic CYP450 mRNA expression was highest for castrated and early vaccinated pigs, and lowest for entire male pigs, suggesting suppression at the transcriptional level of CYP1A2, CYP2A and CYP2E1 by testicular steroids. This did not correspond exactly with protein expression and activities, suggesting other regulations for some CYP450s. The levels of skatole and androstenone in adipose tissue were low in castrated and vaccinated pigs, whereas entire male pigs had elevated levels. Daily weight gain and feed conversion did not differ between groups and income per carcass did not differ between castrated or vaccinated pigs. The income was lower for entire male pigs, due to the additional cost for boar taint analyses and reduced payment for tainted carcasses. After vaccination, the frequency of interactions, both problematic and non-problematic, decreased from the levels of entire males to those of surgically castrated pigs. Under these conditions, early vaccination with Improvac can be used to control boar taint, testicular function and behaviour with unaffected profitability. Thus, the flexibility of vaccination may be extended, implying advantages in terms of animal welfare and sustainability. Keywords: Early immunocastration, male pigs, boar taint, androstenone, skatole, testicular function, hormones, cytochrome P450, performance, behaviour.
Author’s address: Carl Brunius, SLU BioCenter, Dept of Food Science, P.O. Box 7051, 750 07 Uppsala, Sweden E-mail:
[email protected]
Dedication To my family – past, present and future.
To know that one knows what one knows, and to know that one doesn’t know what one doesn’t, there lies true wisdom Confucius
The more you know the less you understand. Lao Tzu
Contents List of Publications
7
Abbreviations
9
1
Implications
11
2 2.1 2.2 2.3 2.4
2.5
Introduction HPG axis Modulation of GnRH function Modulation of GnRH function in cattle Modulation of GnRH function in male pigs 2.4.1 Boar taint and castration 2.4.2 Agonistic GnRH modulation 2.4.3 Immunocastration Adverse effects of GnRH function modulation
13 13 15 17 18 18 22 22 25
3
Aims
27
4 4.1 4.2 4.3 4.4 4.5 4.6
Materials and methods Experimental design (II-V) Biochemical analyses (I-IV) Testicular histology and spermatozoal morphology (III) CYP1A, 2A and 2E1 (IV) Behavioural analyses (V) Statistical analyses
29 29 31 32 32 33 34
5 5.1 5.2 5.3 5.4 5.5 5.6 5.7
Results Antibodies Reproductive organs Hormones CYP1A, 2A and 2E1 Boar taint substances Performance and carcass quality Behaviour 5.7.1 Activity behaviour 5.7.2 Social interactions Skin lesions
37 37 38 40 42 46 47 48 48 49 52
5.8
6 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9
Discussion Antibodies Reproductive organs Hormones CYP1A, 2A and 2E1 Boar taint Performance and carcass quality Behaviour Long-term effect of immunocastration Implications for animal welfare and sustainability
53 53 53 54 55 56 57 59 59 60
7
Conclusions
63
8
Indications for future research
65
9
References
67
10
Acknowledgements
77
List of Publications This thesis is based on the work contained in the following papers, referred to by Roman numerals in the text: I Brunius, C. and Zamaratskaia, G. A modified high performance liquid chromatographic method for simultaneous quantification of skatole and indole in porcine plasma (submitted). II Brunius, C., Zamaratskaia, G., Andersson, K., Chen, G., Norrby, M., Madej, A. and Lundström, K. Early vaccination with Improvac® - effects on boar taint, hormones and reproductive organs. Vaccine (accepted). III Einarsson, S., Brunius, C., Wallgren, M., Lundström, K., Andersson, K., Zamaratskaia, G. and Rodriguez-Martinez, H. (2011) Effects of early vaccination with Improvac® on the development and function of reproductive organs of male pigs. Animal Reproduction Science 127(1-2), 50-55. IV Brunius, C., Rasmussen, K. M., Lacoutière, H., Andersson, K., Ekstrand, B and Zamaratskaia, G. (2011) Expression and activities of hepatic cytochrome P450 (CYP1A, CYP2A, and CYP2E1) in entire and castrated male pigs. Animal (in press, doi:10.1017/S1751731111001674). V Andersson, K., Brunius, C., Zamaratskaia, G. and Lundström, K. (2011). Early vaccination with Improvac® - effects on performance and behaviour of male pigs. Animal (in press, doi:10.1017/S1751731111001200). Papers III-V are reproduced with the permission of the publishers.
7
The contribution of Carl Brunius to the papers included in this thesis was as follows: I
Participated in the planning of the experimental work, solely performed the laboratory work, evaluated the results and was responsible for manuscript preparation.
II Participated in the planning of the experimental work. Participated in the collection of samples and performed some laboratory work. Was responsible for statistical evaluation and interpretation of the results and for manuscript preparation. III Participated in the collection of samples, was responsible for analysis of steroids and participated in manuscript preparation. IV Participated in the collection of samples, performed some laboratory work. Was responsible for statistical evaluation and interpretation of the results and for manuscript preparation. V Participated in statistical evaluation and interpretation of the results and in manuscript preparation.
8
Abbreviations COH CYP EIA EROD FSH GnRH GnRHR HPG HPLC IGF-1 LH LS LW MROD PCR PNPH RIA SE
Coumarin 7-hydroxylase Cytochrome P450 Enzyme immunoassay 7-Ethoxyresorufin O-deethylase Follicle stimulating hormone Gonadotropin releasing hormone GnRH receptor Hypothalamic-pituitary-gonadal High performance liquid chromatography Insulin-like growth factor 1 Luteinising hormone Least squares Live weight 7-Methoxyresorufin O-demethylase Polymerase chain reaction p-Nitrophenol hydroxylase Radioimmunoassay Standard error of the mean
9
10
1
Implications
Immunisation of male pigs against gonadotropin-releasing hormone (GnRH) using a commercial vaccine (Improvac®, Pfizer Ltd.) is an available alternative to surgical castration of male pigs. Standard vaccination involves two injections at least 4 weeks apart, with the second injection up to a maximum of 10 weeks before slaughter. This study has shown for the first time that an earlier than recommended immunocastration gave results comparable to surgical castration and standard vaccination regarding boar taint, behaviour, performance and income per carcass. This implies that the time span of vaccination can potentially be extended and the flexibility of vaccination increased for farmers that find standard vaccination impractical. With a flexible vaccination schedule, there are potential benefits for both animal welfare and sustainability: (i) surgical castration is avoided; (ii) immunisation can be timed to the sexual maturity of the pigs, effectively controlling problematic boar behaviour and; (iii) vaccinated pigs have a higher anabolic potential than surgically castrated pigs. Earlier vaccination is a potential alternative to the currently recommended vaccination schedule. However, early vaccination needs to be more extensively studied under different practical conditions before coming to general conclusions and recommendations.
11
12
2
Introduction
Since the domestication of mammals, humans have sought to modulate their reproductive systems to avoid uncontrolled reproduction, to control problematic behaviour and, in the case of livestock, to obtain production benefits in terms of increased growth, weight and/or fattening and the suppression of oppressive odours. This control has historically been achieved through physical castration, mostly of males, either by surgery, elastration or emasculation.
2.1 HPG axis After the discovery of the hypothalamic-pituitary-gonadal (HPG) axis and its role in regulation of the reproductive system (Fig. 1), control has been sought through agonistic or antagonistic interactions using hormonal analogues or immunisation. Hypothalamus
GnRH
Ovary
Oestrogen Inhibin Oocyte
Pituitary
LH & FSH
Testis
Testosterone Inhibin Spermatozoa
Figure 1. The hypothalamic-pituitary-gonadal axis with key hormones. 13
The hormonal cascade in the HPG axis is initiated by GnRH, sometimes referred to as luteinising hormone releasing hormone (LHRH), a neuropeptide conserved across all species of mammals, which was first identified in the mid 1960s in bovine hypothalamus (Schally & Bowers, 1964). Schally’s group was also first to propose the amino acid sequence (Matsuo et al., 1971). It is a linear decapeptide (Fig. 2) enzymatically produced from the prepro-GnRH precursor in hypothalamic neurons and secreted in a pulsatile manner into the hypophysial portal bloodstream (Thompson Jr, 2000; Millar, 2005; Schneider et al., 2006). After transportation to the gonadotrope cells of the anterior pituitary gland, it binds to the GnRH receptor (GnRHR) which in turn triggers the production of the gonadotropic hormones; luteinising hormone (LH) and follicle stimulating hormone (FSH) (Pawson & McNeilly, 2005). In the female, LH then triggers ovulation and maintains corpus luteum function, while FSH initiates growth of ovarian follicles and subsequent oestrogen synthesis. In the male, LH stimulates Leydig cell production of testosterone, while FSH stimulates spermatogenesis in the seminiferous tubuli. FSH is in both sexes also responsible for the release of inhibin, which together with gonadal hormones imposes feedback regulation of the HPG axis. Whereas the regulation of LH secretion is strongly governed by the GnRH-GnRHR interaction, the regulation of FSH seems to be dependent also on other releasing factors (Li et al., 1998; Padmanabhan & McNeilly, 2001; Schneider et al., 2006).
Gly6 Tyr5 Ser4
Leu7 Arg8
Trp3 His2
Glu1
Pro9
Gly10
Figure 2. Peptide structure of GnRH in the folded conformation adopted for receptor binding. Both COOH- and NH2-terminal domains are highly conserved between natural isoforms, indicating importance for receptor binding and activation. Substitution of Gly6 with D-amino acid enhances activity. D-amino acid substitution of amino acids 1-3, which govern receptor binding and activation, is frequent in GnRH antagonists. Adapted from Millar (2005).
Recently, the neuropeptide kisspeptin was discovered to trigger the release of GnRH through binding with GPR54 receptors on the GnRH neurons. Although the mechanisms of kisspeptin and HPG axis regulation are not yet fully 14
explained, the influence of kisspeptin/GPR54 has been amply reviewed, for example by Dungan et al. (2006), Smith et al. (2006) and Clarke (2011). The kisspeptin pathway can also help explain the negative feedback regulation of gonadal steroids on GnRH release, because whereas GnRH neurons seemingly do not express sex steroid receptors, kisspeptin neurons do. Thus, negative feedback on GnRH secretion by steroids seems to be regulated upstream. Many potential targets exist for the purpose of HPG axis modulation. These targets range from GnRH via gonadotropins to gonadal steroids and even gametes and conceptus. Even kisspeptin has been targeted for research purposes (Kinoshita et al., 2005). The main target has however been GnRH due to its pivotal role in governing the HPG hormonal cascade. The focus in this thesis is on GnRH suppression through prepubertal or early pubertal immunisation. Other targets for immunocontraceptive or immunoneutering purposes have been reviewed elsewhere (D'Occhio, 1993; Meeusen et al., 2007).
2.2 Modulation of GnRH function So far, the main target for control of sexual development and function has been GnRH since it is the key hormone governing reproductive function through the HPG axis. Owing to the conservation of the peptide structure of GnRH, modulation would thus regulate reproductive function in all mammals. Modulation of GnRH function is applied for either of two main purposes: profertility or anti-fertility. An improvement of fertility of domestic animals is mostly obtained through short-term or low dosage administration of GnRH agonists, thereby stimulating luteinisation and progesterone production (Peters, 2005). Anti-fertility methods, on the other hand, are based on reversible suppression of the HPG axis. Long-term fertility control is of interest in humans, domestic and companion animals and wildlife, especially when some consequences of physical castration, such as permanent sterility, trauma, production setback and increased mortality are undesirable. Suppression is achieved by either: (i) chronic administration of GnRH agonists; (ii) administration of GnRH antagonists or; (iii) immunisation against GnRH (Fraser, 1982; Herbert & Trigg, 2005). GnRH agonists are GnRH analogues i.e. natural or synthetic peptides bearing structural similarities to mammalian GnRH (Fig. 2), which interact with GnRHR to produce and secrete gonadotropins. The research and development of agonists has focused on improving receptor-binding and activation and consists largely of substituting L-amino acids with D-isomers (Padula, 2005; Schneider et al., 2006). Short-term and/or low level dosage of
15
an agonist results in a surge of gonadotropins, leading to enhanced gonadal function. Long-term or chronic administration however, after an initial surge of gonadotropins (“flare” effect), causes a down-regulation of GnRHR and desensitisation of gonadotrope cells, in general leading to suppression of gonadal activity (Pawson & McNeilly, 2005). GnRH antagonists are also GnRH analogues, which however do not induce gonadotropin release. Instead they block the receptors for access by endogenous GnRH, while not inducing receptor activity themselves (Padula, 2005; Schneider et al., 2006). The use of earlier generations of GnRH antagonists was associated with significant histamine release. However, later generations have been developed to avoid this side-effect (Padula, 2005). Immunisation against GnRH requires a sufficient amount of antibodies against GnRH present in the hypophysial portal bloodstream, leading to the neutralisation of free GnRH before reaching the receptors in the pituitary (Thompson Jr, 2000). Immunisation can be either active or passive, where in active immunisation the subject is injected with an antigen, against which the body develops antibodies. These antibodies later bind to and neutralise endogenous GnRH if the antigen is structurally similar enough to GnRH. In passive immunisation, however, the subject is injected with readymade antibodies raised elsewhere. While passive immunisation is more selective and reproducible, it also requires large volumes of antisera and does not produce prolonged immunological effects and is therefore of little practical use, especially in livestock (van der Lende et al., 1993). Generating antibodies against GnRH faces two distinct problems: (i) GnRH is a self-peptide and as such is not naturally immunogenic; (ii) GnRH in itself is too small to elicit an immunological response. Antigens must therefore be constructed from GnRH or an analogue conjugated to an immunogenic carrier. The immune system develops antibodies against the conjugate and is thereby deceived into recognising endogenous GnRH as foreign (Thompson Jr, 2000; Herbert & Trigg, 2005). The first developed vaccines used GnRH adsorbed or bound to carriers, such as polyvinylpyrrolidone, bovine serum albumin or glutaraldehyde (Thompson Jr, 2000). Later vaccines have in general focused on synthetic analogues conjugated to carrier proteins, such as bacterial toxoids or ovalbumin (Ferro & Stimson, 1998; Meeusen et al., 2007). Normally, peptidecarrier conjugates are adjuvanted to improve the immunogenic effect. The more effective the adjuvant, the larger is the provoked immunologic response, with a longer lasting effect. However, this normally also results in increased local tissue response or even damage and may not be possible from a practical or legislative perspective. On the other hand, with a less prominent immune
16
system response comes the need of booster injections, to guarantee immunisation or prolonged immunisation (Thompson Jr, 2000; Herbert & Trigg, 2005). There are three main areas of interest for the suppression of GnRH function in animal production: (i) to reduce problematic (aggressive and sexual) behaviour and increase intramuscular marbling in bulls; (ii) to prevent oestrous behaviour and fertility in heifers and; (iii) to control boar taint and problematic behaviour in male pigs (Bonneau & Enright, 1995). Lately, the use of GnRH modulation using analogues has also received attention for the purpose of increasing fertility in livestock, albeit with varied results (Bartolome et al., 2005; Peters, 2005; Brüssow et al., 2011). Although of enormous interest in terms of fertility, production and economy, this particular usage will not be further discussed. There is an obvious advantage in using immunisation rather than administration of analogues for livestock management in terms of residual components in the meat (EC, 1996). Hormonal analogues are administered continuously for maintained effect, whereas vaccines after injection (several weeks before slaughter) are degraded almost instantaneously. Using vaccination, there is thus no risk of hormonally active or disrupting agents remaining in the final meat or milk products.
2.3 Modulation of GnRH function in cattle Bulls Although intact bulls grow more rapidly and have higher feed efficiency than castrated animals, castration of bulls is often performed to control behaviour and enhance the palatability of the meat, through increased marbling and tenderness. Treatment with GnRH antagonists seems to be an effective means of controlling testosterone production in bulls (Jiménez-Severiano et al., 2007), whereas application of GnRH agonists result in increased testicular activity and testosterone concentration, contrary to the general mechanism previously described (Adams, 2005; Jiménez-Severiano et al., 2007). Immunisation against GnRH has been shown to reduce spermatogenesis and steroidogenesis in bulls, resulting in the advantages of castrates discussed above (Price et al., 2003). Moreover, immunised bulls do not suffer from the disadvantages of diminished growth rate and feed conversion to the same extent as castrated animals. However, immunisation against GnRH still remains impractical due to the large variation in individual response to vaccination (Adams, 2005). Undoubtedly, this variation results to a large degree from individual differences in immunocompetence, but vaccine 17
formulation in terms of antigen, carrier and adjuvant is also a large contributor. Novel vaccines may improve immunogenic response and increase the potential for GnRH immunisation treatment as an alternative to castration (Theubet et al., 2010). Heifers Sexually mature heifers show oestrous behaviour and may also become accidentally pregnant. These potential managerial issues may be addressed through modulation of GnRH function. Treatment of heifers with GnRH agonists, contrary to bulls, may result in effective control of the entire HPG axis through the use of long-lasting depot formulations (D'Occhio et al., 2000; 2002). Depot formulations of hormonally active substances in livestock may however constitute a problem, especially in the EU (EC, 1996). Vaccines have suffered from the previously described variation in individual response. However, promising results on the use of recombinant vaccines against GnRH may increase the potential of immunisation for control of heifer fertility (Stevens et al., 2005; Geary et al., 2006).
2.4 Modulation of GnRH function in male pigs 2.4.1 Boar taint and castration
Surgical castration of male piglets is still common practice to avoid boar taint and to reduce problematic behaviours associated with sexual maturity of the boar (Fredriksen et al., 2009; Lundström et al., 2009). Boar taint is an offensive odour pronounced upon heating of meat products from entire male pigs and is caused mainly by two compounds: androstenone, a steroid synthesised in the Leydig cells of the testes analogously to testosterone (Fig. 1) and; skatole, which is bacterially produced from tryptophan in the large intestine (Zamaratskaia & Squires, 2009). Androstenone, although structurally similar to testicular steroid hormones, possesses no hormonal activity but acts as a pheromone. After synthesis, it is absorbed into the blood, after which some of the circulating androstenone is excreted via the salivary glands and, together with other androst-16-ene steroids, induces the mating stance of the sow (Signoret, 1970; Dorries et al., 1995). Skatole, on the other hand, is to the greater part excreted directly via faeces, although a portion gets absorbed into the blood. After circulation, skatole is metabolised in the liver by cytochrome P450 (CYP450) isoforms 1A2, 2A and 2E1 and subsequently excreted via urine (Fig. 3) (Matal et al., 2009; Zamaratskaia & Squires, 2009). Androstenone and skatole do not account for all cases of boar taint reported by sensory analysis. Other substances have therefore been suggested to be 18
important in boar taint. One of these substances, although not as important as androstenone and skatole is indole, a structural analogue to skatole. It is usually measured simultaneously with skatole (Haugen et al., 2011). p-Cresol has also been mentioned as a potential candidate. However, its contribution to boar taint is likely minimal (unpublished results).
a) Androstenone
Uptake
Saliv. gland
Fæces
?
Fat
Urine
Circulation
Metabolism
Blood
Liver
Fat
Excretion
Colon
Mobilisation
Liver
Mobilisation
Production
Blood
Accumulation
Uptake
Metabolism
Accumulation
Testes
b) Skatole
Circulation
Excretion
Production
Urine
Pheromonal activity
Figure 3. Dynamic aspects of a) androstenone and b) skatole partitioning in the pig. Partitioning dynamics are similar for the two substances, except that: androstenone is excreted from the salivary glands and induces the mating stance in the sow; skatole is partly excreted via faeces, and may under poor management be reabsorbed.
Owing to the lipophilic character of androstenone, skatole and indole, deposition from the blood into the adipose tissue can occur (Fig. 3). The extent of deposition determines whether or not meat is tainted. Although sensory evaluation is the only true measure of taint, threshold values for instrumental measurements of androstenone and skatole concentrations in fat are often employed. There is no international consensus on threshold values, but levels of 0.5-1 μg/g fat for androstenone and 0.2-0.25 μg/g fat for skatole are usually reported in the literature (Walstra et al., 1999). The level of skatole in fat is determined by the relationship between formation/absorption on one hand and metabolism on the other. The role of gender in the regulation of CYP450 enzyme activities is well established and the metabolism of skatole by CYP1A2, 2A and 2E1 is generally slower in the boar compared to barrows and gilts. It is likely that the gender-related differences are due to the presence of testicular steroids affecting these enzymes. Androstenone and oestrogens which are produced in increased concentrations in the boar testis (Raeside et al., 2006) were suggested to be particularly important in this modulation (Zamaratskaia et al., 2007; Zamaratskaia et al., 2011). Studies on the interaction between testicular steroids and selected CYP450 have shown an effect both in vitro (Doran et al., 2002; Zamaratskaia et al., 2007; Chen et al., 2008; Rasmussen et al., 2011c) and in vivo (Whittington et al., 2004; Kojima et al., 2008; Zamaratskaia et al.,
19
2008b; Zamaratskaia et al., 2009). Castration thus results in the absence of androstenone formation through the removal of the testes and a subsequent increase in skatole metabolism. It is generally believed that dominance and sexual and aggressive behaviours of the boar are related to high circulating levels of testosterone (Signoret, 1976; Ellis, 1986; Giersing et al., 2000). After castration, testosterone is only produced in limited amounts in the adrenal gland and aggressive and sexual behaviours are as a consequence reduced (Cronin et al., 2003; Rydhmer et al., 2010). Animal welfare and sustainability There are animal welfare issues to consider in raising both barrows and boars. On one hand, the castration procedure is normally performed without the use of anaesthesia or analgesia and most often within the first 7 days of life. This is undoubtedly associated with pain, discomfort and stress, but also an increased risk of infection and mortality (EFSA, 2004; Prunier et al., 2006; Llamas Moya et al., 2008; Fredriksen et al., 2009). On the other hand, boars have in general increased aggressive and sexual behaviour, problems often overlooked in the literature on raising entire males. In feral pigs, boars seem to be divided into dominant, mobile individuals and more sedentary, subordinate individuals (Saunders & Kay, 1991; Hampton et al., 2004). Thus, animal welfare consequences of problematic behaviour related to high testosterone levels are in feral pigs naturally minimised. In a confined space, such as a production site, individuals have no practical means of avoiding conflicts and it is likely that high testosterone and its associated problematic behaviour would negatively affect all three states of animal welfare (physical, mental and telos/naturalness). Even if raising entire male pigs implies disadvantages in terms of animal welfare, there are some important advantages for sustainability. Entire males generally exhibit increased feed efficiency, leaner carcasses and higher nitrogen retention, through the increased deposition of protein compared to barrows (Xue et al., 1997; Bonneau, 1998; Lundström et al., 2009). Moreover, there are some indications that surgical castration may affect the long-term welfare of pigs in terms of increased morbidity and mortality compared to boars (Tielen, 1974; de Kruijf & Welling, 1988; Strøm, 1996; Lessard et al., 2002), which should also presumably be reflected in decreased productivity. Public and political concern about the negative impact of castration on animal welfare is continuously increasing. In Norway, this concern resulted in a ban on surgical castration of male piglets without anaesthesia in 2002 and a full ban on surgical castration effective from 2009 although this was later
20
postponed due to the difficulties in finding viable alternatives. At present, no new date is set for the full ban (Bente Fredriksen, personal communication). In Switzerland, corresponding dates are 2010 and 2015 for the partial and full ban, respectively. In the EU, this concern was embodied in a declaration signed in 2010 by several actors in the European pig sector “committed to a plan to voluntarily end surgical castration of pigs in Europe by 1 January 2018. As a first step, from 1 January 2012, surgical castration of pigs will be performed with prolonged analgesia and/or anaesthesia, if carried out” (EC, 2010). It remains to be seen whether this declaration will practically affect pig management. Alternatives to surgical castration. An overview of the entire range of alternatives to surgical castration without anaesthesia is beyond the scope of this thesis. This was reviewed by EFSA (2004) and later updated by von Borell et al. (2009). However, a few of the more currently viable alternatives include: (i) raising entire male pigs; (ii) surgical castration with anaesthesia and/or analgesia or; (iii) immunisation against GnRH, so-called immunocastration. Still, genetic selection, although not presently a viable option due to problems with fertility and productivity, represents the long-term, sustainable target. Raising entire male pigs for meat production is associated with production benefits, but also problematic behaviours (as described above). This type of production may be combined with various techniques or approaches, such as: (i) slaughtering at lower age/weight to reduce the risk of boar taint; (ii) sorting out tainted carcasses at the slaughter line and/or; (iii) learning to live with boar taint through cultural adaptation or accepting a market gap consisting of consumers sensitive to boar taint. Apart from the aspects of raising boars already discussed, to date there exists no satisfactory online method to detect tainted carcasses. Various analytical techniques for research purposes have been developed (Haugen et al., 2011), but unfortunately they either require advanced sample preparation or are too technically demanding for routine use in an abattoir environment. At present, only a small portion of male pigs are raised intact and only in certain markets, such as the UK, Ireland and some specialty products from southern Europe (Fredriksen et al., 2009). Surgical castration with the use of anaesthesia and/or analgesia is performed to various degrees in some European countries. In Sweden, key actors in the pig sectors have signed an agreement which implores farmers to perform castration in combination with nonsteroidal anti-inflammatory drug (NSAID) administration (SvDHV, 2011). This has in a recent report to the Swedish Board of Agriculture been shown to have only limited effect in suppressing
21
pain and discomfort in piglets, whereas a combination of local anaestetics and NSAID produces a better effect (Hansson et al., 2010; 2011). In Sweden, the right to administer such substances has been extended to animal keepers from previously being restricted to veterinarians (Jordbruksverket, 2011). This procedure has thus been facilitated. It should be noted that the administration of anaesthesia and/or analgesia also results in extra workload/time/cost and involves extra handling of the piglets causing increased stress (Hansson et al., 2011). In Norway and Switzerland, as previously mentioned, practically all male piglets are castrated under anaesthesia, and in Denmark and Germany castration in combination with NSAID was initiated in 2009. In the Netherlands, castration is performed under CO2 anaesthesia although not by law, but after an industry agreement. This procedure has been questioned due to the high level of stress in the early induction phase (Mühlbauer et al., 2010) and cannot be deemed a suitable method from at least the point of view of Swedish animal welfare legislation (Hansson et al., 2011). 2.4.2 Agonistic GnRH modulation
Androstenone production is governed by LH secretion and GnRH release patterns (Fig. 4). Long-acting GnRH agonist administration to mature or pubertal boars was shown to diminish androstenone concentration (Xue et al., 1994; Schneider et al., 1998; Kauffold et al., 2010). However, in some cases the variation in individual response limits this approach from a commercial perspective. Again, the use of depot formulations of hormonally active compounds may limit its commercial use in livestock in some markets (EC, 1996). 2.4.3 Immunocastration
Immunisation against various antigens has been studied. Of the potential targets for immunisation, GnRH seems the most viable candidate (Oonk et al., 1998; Dunshea et al., 2001; Jaros et al., 2005; Zamaratskaia et al., 2008a; Fang et al., 2010; Turkstra et al., 2011). A plausible general approach for antigen formulation would be a modification of amino acids 1-3 i.e. the region responsible for binding and activation of the GnRH receptor (Fig. 2). This would render the antigen hormonally inactive, but still effectively result in specific antibodies against GnRH. Such a peptide could be obtained by deletion of the first amino acid in the sequence, which would render the antigen virtually intact, while effectively hindering receptor binding and activation (McNamara, 1988; 2009; 2010). Other approaches include polymerisation of the GnRH (or analogue) sequence (Oonk et al., 1998) or recombinant GnRH fusion antigens (Fang et al., 2010).
22
Immunocastration of male pigs has been performed commercially since 1998 in Australia and New Zealand using Improvac (Pfizer Inc., formerly CSL Ltd). The said vaccine was in 2009 approved for use in the EU. The mode of action is through active immunisation against GnRH and subsequent control over the HPG axis (Fig. 1 and 4). Immunocastrated pigs are in effect boars until successfully immunised, normally 4-6 weeks before slaughter, and thus have the production benefits of the boar discussed above. After immunisation, the reduced levels of anabolic hormones might however negatively affect performance and therefore lead to increased costs. Such negative effects have not been noticed following standard vaccination (Dunshea et al., 2001; Cronin et al., 2003; Zamaratskaia et al., 2008a). Immunising at an earlier stage might however result in lower growth rate and fatter carcasses. The control over testicular function obtained after immunisation decreases the problematic behaviours of the boar. The painful incision of the standard castration procedure with its associated increased risk of infections is also avoided. Immunocastration thus has a large potential for increased animal welfare. Disadvantages of this procedure include an increased workload/time/ cost compare to entire male pigs, a perceived sense of unnaturalness associated with the modulation (or manipulation) of a “normal” or “natural” hormonal pathway and the risk of accidental self-injection, which after two injections would result in temporary suppression of the HPG axis and associated symptoms also in humans. Accidental self-injections are effectively avoided through the use of a safety vaccinator, “which has a dual safety system providing both a needle guard and a mechanism to prevent accidental operation of the trigger” (EMA, 2010). Nevertheless, vaccine injections should not be performed after an accidental first self-injection or by pregnant women. The exact structure of Improvac has not been divulged; however, it consists of a synthetic peptide GnRH analogue conjugated to a diphteria toxoid carrier protein and is delivered in an aqueous dextran-based adjuvant (EMA, 2010). The manufacturer has shown that their vaccine antigen is hormonally inactive (Clarke et al., 2008) yet effectively produces specific antibodies against GnRH. The vaccine is given subcutaneously as two doses of 2 mL at least 4 weeks apart with the first injection after age 8 weeks and the second 4-6 weeks before slaughter although the risk of boar taint is minimal as much as 10 weeks after the second injection according to the manufacturer (EMA, 2010). What distinguishes Improvac from most of the non-commercialised vaccines is its high efficacy although other vaccine formulations have shown promising results (Fang et al., 2010; Turkstra et al., 2011). Vaccination with Improvac results in active immunisation against GnRH and a subsequent decrease in LH and FSH secretion and testicular function,
23
including androstenone synthesis (Dunshea et al., 2001) (Fig. 4). Later studies have confirmed the efficacy of this vaccine formulation (Jaros et al., 2005; Zamaratskaia et al., 2008a; Pauly et al., 2009). Immunisation against GnRH was shown to increase the activity of hepatic CYP1A, 2A and 2E1 (Zamaratskaia et al., 2009) from the lower levels of entire male pigs similarly to surgical castration (Gillberg et al., 2006; Zamaratskaia et al., 2006). The increased activity is most likely related to a decrease in androstenone and oestradiol concentrations. The inhibition of the skatole metabolising system by these steroids is decreased and skatole concentrations are reduced (Zamaratskaia & Squires, 2009). The vaccine is thus as efficient in controlling fertility, behaviour and boar taint as surgical castration, for which e.g. chryptorchidism must be considered. Deposition in adipose tissue Androstenone
Hypothalamus
Testis
Pituitary GnRH
LH
Skatole
Liver Steroids
Colon Skatole
Skatole metabolism Inhibited by steroids
Figure 4. Relationship between the HPG axis, androstenone production and skatole metabolism. In the boar, production of testicular steroids, including androstenone, in the HPG axis inhibits hepatic skatole metabolism. Androstenone and skatole are therefore deposited in the adipose tissue. In the immunocastrate, the lack of free circulating GnRH arrests testicular steroid production including androstenone. Skatole metabolism is increased due to the lack of inhibition.
The effects of immunisation are supposedly transient and a return of functionality is initialised after antibody titres have decreased below a threshold level (Hilbe et al., 2006; Claus et al., 2008). In fact, the manufacturer indicates effective control over boar taint up to 10 weeks after the second injection. However, it has been shown that the effects of Improvac can be present at least 22 weeks after the second injection (Zamaratskaia et al., 2008c). Moreover, the long-term effect of immunocastration, or even the reversibility, may be questioned if the procedure were to take place at a lower age i.e. if testicular development was hindered at a crucial developmental stage (Setty, 1979; Sofikitis et al., 2008). Although, as mentioned, Improvac is approved for use in many markets, the practical use of this vaccine has been limited. There is at present no routine in Sweden or other EU member states excluding Belgium for how to treat meat from vaccinated animals in the chain from abattoir to consumer. In Belgium, however, the large retailer Colrouyt has adopted a policy where meat from 24
male pigs is exclusively from vaccinated pigs. Moreover, organic producers in Sweden have expressed a desire to use immunocastration (Lars Hellbom, KRAV, personal communication). However, this is at present not perceived as compatible with the EU Regulation on ecological production (EC, 2007; KRAV, 2011).
2.5 Adverse effects of GnRH function modulation As already stated, suppression of GnRH function leads to suppression of the entire HPG axis. Such therapies thus provoke effects related to low levels of gonadal hormones. In female mammals, this translates into symptoms normally related to menopause in humans, such as decreased libido, mood alterations and osteoporosis. In the male, effects include increased risk of obesity, diabetes, cardiovascular diseases and osteoporosis (Saylor et al., 2009). Whether these effects are adverse or not depends on the application and the time-scale of the therapy. In human GnRH therapy for treatment of endometriosis (Batzer, 2006) or cancer (Chengalvala et al., 2003; Montagnani Marelli et al., 2006), most notably prostate cancer (Huhtaniemi et al., 2009), these effects must be considered adverse. However, for immunocastration of male pigs, such effects as decreased libido and mood alterations are in fact desired. Moreover, it is known that GnRH receptors are also present in many extrapituitary tissues, e.g. breasts, gonads, heart and central nervous system (Ramakrishnappa et al., 2005; Rispoli & Nett, 2005). GnRH is thus not only the pivotal hormone regulating reproduction, but seems to also be a neurotransmitter with physiological effects throughout the central nervous system. Suppression of GnRH function, either by analogues or immunisation, can thus potentially alter important physiological functions other than reproductive function (Kirkpatrick et al., 2011).
25
26
3
Aims
This is the first study to investigate prepubertal or early pubertal immunisation of male pigs against GnRH. The rationale was created from the combination of the potential long-term effects of immunological castration together with a desire among some farmers to avoid group treatment of older and heavier pigs (Einarsson, 2006). An aim was also to discuss animal welfare and sustainability aspects of immunocastration, in particular the early vaccination schedule. Aspects that have been investigated include: antibody response size and function of reproductive organs hormones with anabolic effects regulation of skatole metabolising enzymes boar taint substances performance and carcass quality behaviour and long-term effects of early vaccination.
27
28
4
Materials and methods
4.1 Experimental design (II-V) A total of 192 crossbred male pigs (Swedish Yorkshire dams × Swedish Landrace or Swedish Yorkshire sires) from 40 litters were used in this study, comprising two identical trials each with 96 pigs. The study was performed at Funbo-Lövsta Research Station, SLU, Uppsala, in accordance with Swedish regulations for use of pigs. The sires used were randomly selected from sires available for artificial insemination. Piglets within litter were in each trial randomly allocated to four equally sized groups at birth. In the first group, piglets were surgically castrated without anaesthesia before the age of one week. Pigs in the second, early vaccination, group were vaccinated with Improvac (Pfizer Ltd., Sandwich, Kent, UK, 2 mL per injection) when aged 10 weeks (72.5 ± 7.2 days (mean ± standard deviation); live weight (LW) 28.9 ± 7.1 kg) and 14 weeks (100.0 ± 7.1 days; LW 47.3 ± 9.9 kg). Pigs in the third, standard vaccination, group were vaccinated when aged 16 weeks (114.0 ± 7.1 days; LW 58.9 ± 11.3 kg) and 20 weeks (142.0 ± 7.1 days; LW 86.2 ± 14.4 kg). The injections were given just behind and below the base of the ear (200 μg GnRH analogueprotein conjugate/mL). Pigs in the fourth, entire male, group were kept intact throughout the study. No placebo substance was administered. In Papers III and IV, a smaller subset of the population consisting of 8 individuals per treatment group from the first trial was investigated. The growing/finishing period started when the pigs were at an age of 72.3 ± 7.0 days and had an initial LW of 29.3 ± 7.4 kg. Each pen held 8 pigs from only one treatment group. All pigs were fed restrictedly with the same commercial diet (12.4 MJ ME per kg, digestible CP 13.5%) twice a day according to the standard feeding regimen for growing/finishing pigs in Sweden (Andersson et al., 1997). Pigs were weighed individually at the start of the study then fortnightly until their final weighing one day prior to slaughter. 29
Feed consumption was recorded on a daily basis and feed conversion ratio was calculated pen-wise. During the growing/finishing period, two surgically castrated, two early vaccinated and one entire male pig died or had to be euthanised for reasons not related to the study. Blood samples were taken by jugular venipuncture from all pigs on four occasions: before the first Improvac injection of both vaccination regimens i.e. at 10 and 16 weeks of age (72.8 ± 7.1 and 114.8 ± 7.1 days, respectively), thereafter at an age of 22 weeks (156.8 ± 7.1 days) and one day prior to slaughter (25 weeks, 177.0 ± 8.3 days). To obtain plasma, blood samples were collected into heparinised vacutainer tubes, separated by centrifugation at 2000 g and stored at –80°C until analysis. To obtain serum, blood samples were collected in vacutainer tubes without heparin, separated by centrifugation at 2000 g and stored at –20°C until determination of antibodies against GnRH. Slaughter was performed in a commercial Swedish abattoir according to standard procedure on two occasions per pen at an average age of 25 weeks (178.0 ± 8.3 days) and an average LW of 119.3 ± 10.2 kg with the four heaviest pigs slaughtered on the first occasion. All pigs were mixed with unfamiliar pigs during transport (350 km) and lairage, to simulate normal transport and slaughter conditions. Before cooling, carcass weight was recorded and lean meat content was evaluated with a Hennessy Grading Probe. The amount of abdominal fat was recorded. Samples of adipose tissue were taken from the neck region of the carcass and kept at –20°C until analysis. Testes and bulbourethral glands were removed and dissected from extraneous tissue. The lengths of both bulbourethral glands were measured to record the average length, and testes were weighed as pairs. Details of sampling of proximal and distal testes and cauda epididymal content are described in Paper III. Additionally, skin lesions were visually recorded at slaughter by one observer using a 6-point scale (0: no visible skin damage; 5: very highly damaged skin). These records were used to classify the pigs into two groups: without skin lesions or with skin lesions. Pigs with only a few light skin lesions (score 0 or 1) were classified as ‘without’. Vaccination schedules and sampling are graphically represented in Fig. 5. Daily lean meat growth from start of the experiment to slaughter was calculated using a formula of our own invention: % lean × (carcass weight initial weight × 0.72) / days in experiment, with the value 0.72 representing a hypothetical dressing percentage at start. The income per carcass was calculated on the basis of carcass weight and estimated lean meat content, according to prices from the Swedish co-operative slaughterhouse, contract note November 2010. For entire male pigs, payment was reduced owing to
30
boar taint analysis and tainted carcasses according to standard procedure. Prices were converted from SEK into EUR (1 SEK = 0.11 EUR). The study was approved by the local Ethics Committee on Animal Research, Uppsala, Sweden, ensuring compliance with EC Directive 86/609/EEC for animal experiments. Early vaccination 1 Age (week)
10
Slaughter
Standard vaccination 2
14
Blood
1
16
2
20
Blood Behaviour
Behaviour
22
25
Blood
Blood
Testes BU gland Liver Fat Lesions
Behaviour
Figure 5. Time schedule for experimental design and sampling.
4.2 Biochemical analyses (I-IV) Skatole in plasma was quantitated by HPLC as described by Zamaratskaia et al. (2004a), modified and validated by Brunius and Zamaratskaia (Paper I). Androstenone, skatole and indole concentrations in fat were measured using HPLC as described by Chen et al. (2006). Antibodies against GnRH were measured by Frontage Laboratories Co., Ltd., Shanghai, P.R. China, using an in-house validated ELISA method (SHAM-043-R0). Testosterone concentration in plasma was measured using a commercial RIA kit (TKTT, Diagnostic Products Corporation, Los Angeles, CA, USA), according to the manufacturer’s instructions. Insulin-like growth factor 1 (IGF-1) concentration in plasma was measured using a commercial EIA kit (DSL-10-2800, Diagnostic System Laboratories, Webster, TX, USA) in accordance with the manufacturer’s instructions. Oestradiol-17β concentration in plasma was measured using a commercial EIA human salivary kit (1-3207, Salimetrics, State College, PA, USA), through an adaptation of the manufacturer’s protocol (Paper II). Chemical analyses were performed in duplicate, except oestradiol and IGF-1 analyses for the surgically castrated and vaccinated groups, for which one analysis per sample was performed. Surgically castrated pigs were excluded from anti-GnRH analyses since it is known that they are at the same low level as entire male pigs (Zamaratskaia et al., 2008a).
31
4.3 Testicular histology and spermatozoal morphology (III) Testes samples were paraffin-embedded and conventionally sectioned and stained with haematoxylin and eosin for histological analysis. The tubular diameter in each specimen was measured manually using an ocular micrometer when examining the slides at 100× magnification. At the semen laboratory, sperm morphology was evaluated by experienced laboratory assistants in wet formol saline-fixed preparations (Bane, 1961) and in air-dried smears stained with carbolfuchsin-eosin according to the method described by Williams (1920) and modified by Lagerlöf (1934). In the wet smears, 200 spermatozoa, when possible, were checked under a phase contrast microscope (1000× magnification). All abnormalities on any given spermatozoon were counted and the overall frequencies were classified according to Bane (1961). For a more detailed examination of the sperm head, when possible, 500 spermatozoa were checked in each stained smear under a light microscope at a magnification of 1000×. Sperm head morphology was classified according to Lagerlöf (1934). The morphological abnormalities were expressed as a percentage of the total number of counted spermatozoa.
4.4 CYP1A, 2A and 2E1 (IV) Microsomal preparation Microsomes were prepared by a calcium aggregation method using TRISEDTA homogenisation buffer (Rasmussen et al., 2011a). Protein concentrations in the microsomes were assayed with a commercially available kit (Bio-Rad Laboratories Inc., Hercules, CA, USA) according to the manufacturer’s instructions. Microsomes were diluted to a protein concentration of 4 mg/mL and stored at –80°C until use within 1 month of preparations. Real-time RT-PCR Isolation of mRNA, reverse transcription and PCR were performed as described by Rasmussen et al. (2011d). Relative mRNA expression was calculated from the obtained Ct values and normalised against mRNA expression of GAPDH. The average expression in the group of entire male pigs was arbitrarily set to 1. Western blot Western blotting was performed according to Rasmussen et al. (2011b). The antibodies used were all raised against a human epitope but have been shown
32
to be specific against porcine cytochromes (Rasmussen et al., 2011d). Relative protein expression was quantified using ImageJ (version 1.43u, NIH, USA) and expressed relative to the average of the group of entire male pigs. Microsomal enzyme activity The CYP-catalyzed activities of the following enzymes were measured: 7-ethoxyresorufin O-deethylase (EROD; CYP1A1/1A2); 7-methoxyresorufin O-demethylase (MROD, CYP1A2); coumarin 7-hydroxylase (COH; CYP2A) and; p-nitrophenol hydroxylase (PNPH; CYP2A/2E1). EROD and MROD activities were measured as described by Zamaratskaia and Zlabek (2009), and COH activities as described by Zamaratskaia et al. (2009). The activity of PNPH was measured using a simplified HPLC-based method (Paper IV).
4.5 Behavioural analyses (V) Activity behaviours and social interactions of pigs from all groups were studied by direct observations on three occasions per pen: at ~13 weeks (the week before the second injection of the early vaccinated pigs), at ~19 weeks (the week before the second injection of the standard vaccinated pigs), and at ~22 weeks (2-3 weeks before slaughter) (Fig. 5). The observations were recorded by one observer standing outside the pen. Observations did not start until the pigs were accustomed to and no longer seemed to pay attention to the observer. All observations were performed between 10:00 and 15:30 i.e. outside feeding time. Nine observation rounds per pen were made in a consecutive pen order to distribute the observations of pens equally over time, and each pen was studied for a 10-min session per round. The observations consisted of two kinds of sampling and recording: instantaneous scan sampling of activity behaviours, and continuous recording of frequencies of social interactions (Fig. 6). Instantaneous scan samplings of activity behaviours were performed at the beginning and end of each observation round. Between these scan sampling observations, frequencies of social interactions were recorded for a total of 8 min. Thus, one observation day gave 18 instantaneous scan samples and a total of 72 min of social interactions per pen. The behavioural observations were performed over a total of 4 days per occasion. Min: 0
1
Scan
9 10 Social interactions
x 9 rounds
Scan
Figure 6. Time schedule of an observation round for behavioural analysis.
33
The definitions of the behaviour parameters are presented in Table 1 and 2. Contact during scan sampling included aggressive as well as non-aggressive behaviour. In the continuous frequency recording of social interactions, problematic (aggressive and mounting) and non-problematic (sniffing, pushing, crowding, manipulating pen mate and playing) interactions were recorded separately. Table 1. Definitions of behaviour parameters during scan sampling. Behaviour parameter
Definition
Resting
Lying or sitting down
Standing
Standing, walking or running
Eating
Head in the trough or waiting for feed beside the trough or drinking
Contact
Touching another pig in some way including mounting
Table 2. Definitions of behaviour parameters during continuous recording. Behaviour parameter
Definition
Non-problematic Sniffing
One pig sniffing at another pig or nose to nose contact
Pushing
One pig pushing or nibbling or lifting another pig
Crowding
Two or more pigs pushing each other to reach feed or water
Manipulating pen mate
One pig has another pig’s tail or ear in its mouth
Playing
One pig playing (jump, carry straw or run)
Problematic Aggressive
Two or more pigs fighting or giving or receiving head-knocks or bites
Mounting
One pig is mounting another pig
4.6 Statistical analyses Data were analysed with SAS 9.2 (SAS Institute, Cary, NC, USA). For the analysis of oestradiol and IGF-1 in plasma, data were obtained and analysed only from the first trial. For all other analyses, data were combined from the two trials. In Papers III and IV, data were analysed for the smaller subset of the population, consisting of 8 individuals per treatment group from the first trial. The effect of treatment on testes weight, bulbourethral gland length, indole, skatole and androstenone in fat and oestradiol and IGF-1 in plasma, performance, carcass quality and skin lesions recorded at slaughter was evaluated with the MIXED procedure. The model included the fixed factor of treatment (surgical castration, early vaccination and standard vaccination, and
34
entire males) and the random factors of trial, pen and litter. When analysing daily weight gain during suckling, surgically castrated pigs were compared with all entire pigs i.e. also pigs intended for vaccination. To evaluate the effects on variables measured repeatedly in plasma (testosterone and skatole) and serum (antibodies against GnRH), the model included the same factors as above with an additional repeated statement (by individual, unstructured covariance structure). Main effects at different time points were evaluated using the slice statement. The concentrations of skatole, testosterone, oestradiol and IGF-1 in plasma, antibodies against GnRH in serum and concentrations of indole, skatole and androstenone in fat were log-transformed prior to statistical analyses to normalise residual distributions. The effect of treatment on mRNA and protein expression and enzymatic activity was evaluated using the MIXED procedure on rank-transformed data, with treatment as fixed factor. Pair-wise comparisons between treatments were obtained through the pdiff option. Spearman correlations between measured variables were calculated for all pigs. Activity behaviours were recorded as the percentage of pigs performing a particular behaviour at each sampling occasion (9 rounds * 2 scan samplings) and were analysed as percent of time. The social behaviour was recorded as the total number of interactions performed per pen and hour at each sampling occasion (9 rounds * 8 minutes). Pen was the statistical unit for all behaviour analyses. These parameters were evaluated within each observation occasion with the MIXED procedure. The model included treatment as fixed factor and trial as random factor. For comparisons over time, analogous evaluation was performed within treatment, with occasion as fixed factor and trial as random factor. Activity behaviour data were transformed via ‘arcsine (square root)’ and social interactions via ‘square root’ before statistical analysis to normalise residual distributions. The impact of treatment on skin lesions recorded at slaughter was tested as a logistic regression using a binomial distribution with a logit link function. These analyses were done with the GENMOD procedure and the model included the effect of treatment and trial.
35
36
5
Results
5.1 Antibodies Titres of antibodies against GnRH were detected only in low concentrations in serum from entire male pigs (Fig. 7). For both groups of vaccinated pigs, antibodies were at the same concentration as for entire male pigs until the second vaccine injection. After the second injection, the concentration of antibodies increased rapidly and then gradually decayed. All vaccinated pigs had increased concentrations of antibodies at the end of the trial. 2000
Early vaccination
b
Standard vaccination
Anti-GnRH antibodies (units/mL)
1800
Entire male pigs
1600
c
1400 1200
c
1000 800
b
600
b
400 200
a
a
a
a
0 10 weeks
16 weeks
22 weeks
25 weeks
(p
